isolated Tx - no service neutral

[GITRDUN]

My electrician was in the shop today to wire up a CNC machine for us and while i had his ear i asked him if there was a reason we do not seem to have a neutral line coming from the utility lines outside. He told me that our 480Tx makes its own neutral and that we do not need the neutral from the utility. Makes sense to me i guess but then what do i know. But i have had a hunch for some time that the lack of a neutral is what has been causing our machine problems. Let me explain our power setup and the problems we have had for a long while now.

1. The 3 phase utility service is connected to a 480Y transformer which creates its own neutral and does not have a neutral coming from the utility. The 3 phase plus neutral coming from this 480Tx goes into a large main breaker where it then branches out into two feeds. One feed has the 3 phase plus neutral fed into an over head bus bar running through the shop. This feeds 480v 3 phase power to some of our cnc equipment.

2. The second feed from the 480y has 3 phase with no neutral going into a 208Y tx which creates its own neutral. Coming out of the 208tx is 3 phase 208y with a neutral. The 3 phase is fed into a main breaker. From the main 208 breaker the 3 phase then goes into a junction box which has a seperate bar for each of the three phase legs. All power to each individual breaker box down line whether it be single phase, dual phase or three phase is connected to the 3 phase bars in this junction box. The neutral wire from the 208tx runs into the junction box and is connected to the buildings ground rod and also all other ground wires going to machines and each breaker box down line. Our single phase is derived from the 208.

How our cnc equipment is wired up -
Most all of our machines are powered from the 208V. They have three phases coming from a breaker and also a ground wire connected to the building ground bar, the power wires then go into a disconnect switch on the back of the machines and the ground wires connect to a ground terminal inside the machine cabinet. All machines then pass the three phase into a built in transformer and after which it branches out inside the machine to numerous electronics boards and motor drivers. When powering machines off we turn the controls off and then switch off the disconnect on the back of the machine so all CNC machines are physically disconnected from the grid when not being ran.

Here is our nightmare problem that has eluded us thus far almost exclusively on our 208V equipment. Every couple weeks or so we will have a CNC machine or two blow out a control board when the machine is switched on in the morning. I have a tech man who is capable of fixing our electronics boards and i have enough problems to keep him busy 40hrs a week every week. This ONLY happens in the mornings when a machine is switched on and powered up. It is rare for our machines to blow a board or fault out during operation, if the machine powers up in the morning they run all day no problem. Just yesterday we had 4 out out of 5 machines trip the machines main power breaker built into the machines when they were switched on in the morning. We simply switch the power off, flip the breaker up and power them back on and they work fine. This happens quite often, and when it does its usually not just one machine that does it its multiple machines all in the same morning. When this happens we check voltages line to line and line to neutral and have never found anything out of whack. I have used recording meters on the main lines and even on machines themselves before and could not find anything wrong. I have had an electric consultant company come in and meter our power briefly. They didnt find anything and their recommendation was to rewire the plant. I cant afford to spend $50k rewiring the whole plant to try and fix the problem if i dont know its going to fix our problem. We are a small business and with the economy the way it is its all we can do right now to keep our honest hard working employees employed. I would certainly spend the money fixing the problem if anyone knew what would fix it. But throwing a ton of money at something just hoping it fixes it isnt an option right now unfortunately. I am seeking sound advice from people who certainly know more about this sort of thing than i do and am hoping someone maybe has seen this type of problem before.

 

[petersonra]

There is no way for anyone here to diagnose this kind of thing from a brief Internet forum post.

My suggestion is to get a competent electrician or engineer to come take a look. There are a lot of power system consultants who are clueless when it comes to diagnosing just why this kind of thing is happening, it is just not their field of expertise.

the lack of a neutral coming from the utility is unlikely to be much of an issue given the setup you describe. I am not sure that rewiring the shop will solve anything either.

have you discussed this with the manufacturer of the CNC equipment? preferably a step or two above the typical CNC tech that seem to think you can solve all CNC problems with a ground rod.

since it happens mostly when you power up a machine, I would give strong consideration to not powering them off at night until you figure out what is going on.

 

[GoldDigger]

Connecting what appears to be a 480 delta service (possibly center grounded but no neutral from utility) to the input of a delta/wye transformer will "create" a neutral on the output side, in the same sense that it creates the three hot phase lines.
It is standard practice to connect such a wye supply to a delta wound transformer (no neutral going in) to produce 208wye output.
Nothing seems out of the ordinary based on what little information you have given. The bonding of the various output neutral wires is a potential issue, but you have said nothing to indicate one way or another whether it was done correctly.

As petersonra stated, your problem is not something that can be diagnosed remotely.
Note, however, that opening the disconnect or breaker and immediately closing it again can cause surges in motor equipment and even transformers and should be avoided. Opening the disconnect long enough for any running motors to coast to a stop will be a less stressful procedure.
You did not describe a neutral wire going to the CNC machines. That is probably normal, but should be checked against the specifications of the machines just for completeness.

 

[GITRDUN]

Connecting what appears to be a 480 delta service (possibly center grounded but no neutral from utility) to the input of a delta/wye transformer will "create" a neutral on the output side, in the same sense that it creates the three hot phase lines.

Im sorry if i mistyped or something but it is a 480Wye transformer feeding a 208Wye transformer.

Machines do not require nor have a connection for a neutral wire, only a connection for the ground wire. I have spoken with one of the machine tool manufacturers and they really disnt have any suggestions.

I know this is a long shot probing for explanations via an internet site but you never know, surely somebody out there has seen this same issue before. Cant hurt to throw it out there.

 

[petersonra]

Im sorry if i mistyped or something but it is a 480Wye transformer feeding a 208Wye transformer.

normally these kind of distribution transformers are delta on the input side and wye on the output side.

nothing you are describing seems out of the ordinary other than the troubles you are having with boards blowing up.

 

[GITRDUN]

normally these kind of distribution transformers are delta on the input side and wye on the output side.

Ah, ok, i got it.

 

[GITRDUN]

Also i do realize switching a breaker on and off quickly can cause problems due to motors and also for the capacitors and things. Its not an instant on and off flip of the switch. We give it a brief time before switching back on if it must be turned back on.

 

[petersonra]

Also i do realize switching a breaker on and off quickly can cause problems due to motors and also for the capacitors and things. Its not an instant on and off flip of the switch. We give it a brief time before switching back on if it must be turned back on.

Given the type of equipment involved it seems unlikely that the switching speed of the breaker is the issue.

What does the CNC manufacturer say about this problem?

While the guy was doing the power quality test did anything trip or fail?

 

[GITRDUN]

The CNC tech i spoke with awhile back really didnt know what to make of it.
And no nothing failed while they were recording on the lines. I wasnt impressed with the service from the electric consultants. The boss man showed up to put recording devices on and poke around for 15 minutes. Then later sent a younger fellow in to pick up the equipment and sent me a quote for rewiring the plant. They werent interested in chasing an unusual problem IMO.

 

[petersonra]

The CNC tech i spoke with awhile back really didnt know what to make of it.
And no nothing failed while they were recording on the lines. I wasnt impressed with the service from the electric consultants. The boss man showed up to put recording devices on and poke around for 15 minutes. Then later sent a younger fellow in to pick up the equipment and sent me a quote for rewiring the plant. They werent interested in chasing an unusual problem IMO.

These intermittent problems are not a lot of fun.

 

[GITRDUN]

Here is a rough diagram of our power flow. If anyone sees something wrong with this please point it out. I will have the electrician back in tomorrow and would like to discuss with him more about possible issues with our power distribution if there are any.

 

[fmtjfw]

Check the voltage in the morning. Is it "early" in the morning? It is possible that the feed is unusually unloaded and prodinding a higher than expected voltage? Grasping at straws here.

 

[petersonra]

Here is a rough diagram of our power flow. If anyone sees something wrong with this please point it out. I will have the electrician back in tomorrow and would like to discuss with him more about possible issues with our power distribution if there are any.

If the diagram is correct there is probably at least one code violation WRT how the ground rods are connected in, but nothing that jumps out at me and says that is what your problem is.

It is also required that the neutral be run with the power conductors and your diagram shows it is not.

 

[GITRDUN]

The 208 Tx neutral is connected onto a grounded lug in the first box and then connects to a ground bar inside the junction box which is grounded with a ground rod. Every breaker box down line of that junction box has the ground wire running to it with the power wires. Are you saying the neutral should be kept seperate? Im not understanding what you are saying exactly.

Here is a nugget for you. Tuesday morning i had 4 machines that had their main power breaker inside the machines tripped when the operator turned the machines on. Same old same ole. We power off the machines, trip the breaker back in and power back on and they all ran fine. This morning i came in early and checked every single machine before any machines were powered on and found 5 machines that had their main power breaker tripped. This is before any of them were turned on mind you. 4 of them were the same machines as the ones from Tuesday. The one out of the five that wasnt tripped Tuesday had an axis motor drive board blown when we turned it on. So there is no question that whatever is happening is tripping the breakers inside the machines after we have gone home and with the machine physically disconnected from the power wires. I feel stupid explaining this to people sometimes because i know it makes no sense, i have had people flat out tell me they dont believe me. It is what it is. Someday we will get it figured out, or move.

 

[petersonra]

The 208 Tx neutral is connected onto a grounded lug in the first box and then connects to a ground bar inside the junction box which is grounded with a ground rod. Every breaker box down line of that junction box has the ground wire running to it with the power wires. Are you saying the neutral should be kept seperate? Im not understanding what you are saying exactly.

generally, the neutral wire can only be connected to ground at one point. that would be where the grounding electrode conductor is connected (the GEC is the wire that goes to the grounding electrode - a ground rod in your case).

your diagram appears to show that the neutral conductor is not being run with the three phase conductors. this might just be how you drew it up, but the code requires that the neutral be run with the hots and not separately.

 

[rt66electric]

neutral and grounds on other machines are not seperated?

neutral and grounds on other machines are not seperated?

Here is a straw to grasp at....

neutrals and grounds should be seperate at each individual load... power runs in a CIRCUIT in the black(red,yellow, purple,blue,Etc) through the load and BACK through the WHITE wire ,,, NOT THE GREEN one,

if a machine (load) has the white and green tied together on the other end off the shop, then power(amperge ) returns back to the transformer however it wants to (sometimes through the green wire on the bridgeport mill) .

look at the control voltage transformer in the cabinet,
normally there is a 208 to 120 or 24v observe L1 L2 connections,
observe X1-X0 I remeber corrcect the X0 connection should be tied cabinet ground, the XFMR secondary X1 should be 120 to GND and X0 should be 0volts.... if you are reading Appox 70v and 40v respectfully , then check X0 connection


I would look at machines with large loads and verify that the amperage on the power wires are close to the neutral load.

check all sub-panels and isolate neutrals and grounds, and remove the jumpers.

Another rare occurance is the loaded line wires and the loaded return wires(neutral/green) are in seperate conduits creating a magnetic tranformer(hysteresis effect) and the sudden inrush of power(amperage) is transformed into a voltage spike.

I remember a "BRIDGEPORT official memo" that recomended ..
install a ground rod at the point of the machine and remove the ground wire from the line power.
This "fix" worked well to save control boards, but, the lineworkers would often get a static-shock every time they approched the machine, depending on which type of coolant they used.. ( I tried to convince them to put a nail in the sole of their boots (they did not fall for that gag))


Do you have spotwelders or such in use at the time??

Put you amprobe around the green wires.

There is probably a fault in a machine somewhere down stream.

Check for proper netrual connection (the following is not the safe for the inexperianced) turn machine off , remove and cap the neutral wire, (careful, its loaded) turn machine on, if it runs, then the machine has a neutral to ground fault. remove machine from service
otherwise. the machine is probably good
turn machine off reconnect neutral.

 

[GITRDUN]

Our machines are all three phase and use only the three power wires and a ground wire. They all have their own transformer built into them where their single phase loads are powered from.

 

[gar]

140808-2100 EDT

GITRDUN:

Indicate the CNC manufacturer's name, what is the internal circuit in ths CNC. I assume the CNC is a delta load. This means any 3 phase induction motors are delta across a delta source, the DC power supply for spindle and servo motors is delta and thus floating relative to machine frame (ground), all electronics are supplied from an isolation transformer and referenced to machine chassis. The isolation transformer is probably a 60 Hz isolation transformer followed by a switching power supply that further has a high frequency isolation transformer.

Internal to the machine there may be some type of 60 Hz autotransformer between the delta source and the machine delta loads. If this was tapped wrong, then that would likely produce overvoltage on the large DC power capacitors and cause the machine to flag a DC bus overvoltage and trip something.

There will be power switching electronics that float off of ground associated with the spindle motor and servos. These are isolated with optical couplers or pulse transformers.

You have not indicated what electronic components fail, this is important. Provide this information. A typical design for the DC bus is about 325 V.

You have a somewhat repeatable type of failure, and on duifferent machines. Provide more specific information on what and where these failures occur.

If there is really an input breaker in the machine, and not fuses, then it is strange the breaker trips. This would imply a very high surge current at turn on. I would normally expect a disconnect, fuses, and a master contactor (CRM), rather than a breaker. Over a 20 year period there are very few times the fuses on our CNC machines have blown. The great eastern blackout was one such time, and not all the fuses blew.

.

 

[GITRDUN]

Our lathes are all Okuma brand, our mills are all Haas. I couldnt say whether the machines "load" is delta, but our power setup is 208Wye. The electronics that typically fail are mostly the axis motor driver boards and the spindle motor drive boards. The computer logic boards and memory boards never fail. Almost every drive board in these has fuses and also on board power supplys, sometimes the fuses blow, sometimes they dont. The components on the boards that typically fail are capacitors, volt regulators, and transformers.

After the occurances this week there is no question that this problem is getting into the machines while they are switched off, physically disconnected from the 3 phase power wires. IMO this means that whatever gremlin causing this is coming in through the ground wire or on the conduit.

Today i instructed all employees to switch off every breaker inside of the machines cabinets they can find after they power off a machine. Then switch the breakers back on the next day before powering up a machine. Might or might not help but it cant hurt. The one constant is the machines having breakers tripped off when we come in to work occasionally. I swear im about to call the ghostbusters. This is just weird.

 

[GoldDigger]

Our lathes are all Okuma brand, our mills are all Haas. I couldnt say whether the machines "load" is delta, but our power setup is 208Wye. The electronics that typically fail are mostly the axis motor driver boards and the spindle motor drive boards. The computer logic boards and memory boards never fail. Almost every drive board in these has fuses and also on board power supplys, sometimes the fuses blow, sometimes they dont. The components on the boards that typically fail are capacitors, volt regulators, and transformers.

After the occurances this week there is no question that this problem is getting into the machines while they are switched off, physically disconnected from the 3 phase power wires. IMO this means that whatever gremlin causing this is coming in through the ground wire or on the conduit.

Today i instructed all employees to switch off every breaker inside of the machines cabinets they can find after they power off a machine. Then switch the breakers back on the next day before powering up a machine. Might or might not help but it cant hurt. The one constant is the machines having breakers tripped off when we come in to work occasionally. I swear im about to call the ghostbusters. This is just weird.

If you misguidedly installed local ground rods for each machine and used them instead of connecting to the common EGC wiring, that could cause all sorts of harmful symptoms, but most often when there are nearby lightning strikes or equipment that puts a high objectionable current through the ground system or through the earth itself. (Are you next door to a POCO substation, for example?)